Influenza transmission and its resulting morbidity and mortality are of great concern to society. Strategic intervention may greatly reduce these factors, but the effectiveness of an intervention depends on public adherence, and, more generally, on individual decision making in response to actual and perceived health risks. This project aims to define optimal intervention strategies and policies that significantly improve intervention adherence for both epidemic and pandemic influenza outbreaks. To meet this objective, we will integrate knowledge and methods from epidemiology, mathematical modeling, economics, game theory, and experimental psychology. Contact patterns, which are often dynamic and highly variable, fundamentally influence the spread of disease. These patterns change as individuals make decisions to be vaccinated, accept treatment, take hygienic precautions, or avoid work, school, or public spaces. We will develop new epidemiological models that explicitly consider individual-level perceptions and decisions and their impacts on the contact networks underlying influenza transmission. These models will capture the evolutionary dynamics of influenza, including antigenic drift and the emergence of antiviral resistance. We will apply game-theoretical methods to these models to evaluate different influenza intervention strategies, including vaccination, antiviral-based interventions, and non-pharmaceutical interventions, and to identify strategic opportunities for improving adherence through informational and incentive programs that change individual perceptions and decisions. The contact patterns and psychological components of the models will be based on Bayesian analysis of census data, workflow and recreational mobility data, and real-time influenza surveillance data, as well as on survey studies that evaluate public knowledge and perceptions about the disease. The latter will also provide information on adherence behavior, contact patterns, and the impact of real-time influenza-related decisions on these patterns. Integrating realistic, perception-driven individual-level decision making into epidemiological models will facilitate the evaluation of interventions and the development of strategies to improve adherence both in epidemic and pandemic outbreaks of influenza.